claimkeeper
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
claimkeeper
Keep what your product says about itself consistent — across the README, docs, landing page, pitch text, demo script, voiceover, SVGs and the demo video — while coding agents keep changing things.
You're building a hackathon project with Claude Code. The README says "everything runs on-device", the landing page says "100% private", the voiceover says "no audio leaves your laptop", an SVG says "on-device", and the Remotion video shows it in scene 2. Three sessions later the agent adds cloud sync, updates one file, and your submission contradicts itself.
Git is the tool humans and agents share for code. claimkeeper is the tool the agent uses for claims; you watch it through git diff .claims/, a status line and an activity log.
a claim is an id + one sentence of what it means (
on-device: "All processing runs on the user's device; no audio leaves it.")each claim knows its places: files and exact wording, each place in its own words, short or long
generated outputs (ElevenLabs audio, SVG exports, Remotion renders) remember what they were made from
The tool does the deterministic work (store, find, fetch, rewrite, fingerprint, report). Claude does the judgment (does this contradict? what wording fits here? does the picture still say it?).
What it catches
Situation | What you get |
Lines moved, text reflowed, wrapped in tags | Found anyway (exact → normalized → fuzzy). No false alarms. |
Wording at one place edited |
|
File renamed / deleted |
|
Code that backs a claim changed |
|
Claim meaning changed | Revision bump; every place and output not updated is |
A place contradicts a claim | Registered as |
Voiceover script edited | Audio |
Edits by | Caught by the end-of-turn check and the next session start |
You want the old meaning back |
|
The agent never regenerates media on its own. It tells you what's out of date and asks.
Related MCP server: contradiction-mcp
Install
Requires Node.js 22 or newer (tested on 22 and 24; development uses 24). No runtime dependencies.
claude plugin marketplace add md-abid-hussain/claimkeeperclaude plugin install claimkeeper@claimkeeperOr try it from a checkout without installing: claude --plugin-dir ./plugin.
Codex, Cursor or another agent? Any agent that can use MCP servers works. Tell it:
Install claimkeeper in this project by following https://github.com/md-abid-hussain/claimkeeper/blob/main/INSTALL.md
INSTALL.md walks the agent through registering the MCP server, adding the instructions, turning tracking on and (with your yes) a git pre-commit check. Without Claude Code's hooks, the agent runs the checks itself as its instructions say, and the pre-commit check catches what slips through at commit time.
Turn it on per project
Installing does nothing by itself; hooks stay silent until a project runs init.
/claimkeeper:initNew project (claims first): you name the 3–7 things the product must always say; they become the spec. Places are registered as Claude writes them, and the README, pitch and demo script are written from the claims.
Existing project: a read-only
claim-scoutsubagent reads your README, landing page, scripts, compositions and SVGs in its own context and returns a short proposal with contradictions. You approve; Claude registers.
/claimkeeper:toggle off # quiet for you only (git-ignored setting); "on" to resume
/claimkeeper:toggle relaxed # report problems at the end of a turn, don't block
/claimkeeper:toggle strict # fix or acknowledge new problems before finishing (default)
/claimkeeper:toggle quiet # no end-of-turn check and no new-file suggestions; per-edit warnings only
/claimkeeper:audit # full review (claim-auditor subagent) before you submit or recordWatching it work
You don't have to ask the agent to test anything. Build as usual and look from the side:
node <claimkeeper>/plugin/dist/cli.js log # what the agent and hooks did
node <claimkeeper>/plugin/dist/cli.js status # counts
git diff .claims/ # what changed in the claims since your last commitStatus line (add to ~/.claude/settings.json), showing claims ✓13, claims ⚠2 · 1 pending or claims off:
{ "statusLine": { "type": "command", "command": "node <claimkeeper>/plugin/dist/cli.js statusline" } }Context budget
Everything the agent sees costs context, so claimkeeper discloses progressively:
When | What the agent sees |
Always | a one-line skill description; 7 tool names (schemas load on demand) |
Session start | one line: |
Reading a file with claims | one line, once per file per session: |
Writing claim wording / an output's input | only when something drifted or went stale |
End of turn | only new problems or meaning changes, top 5; blocks at most once each |
Claims skill | under 500 words (a test enforces it), loaded when product-facing text or media is touched; media and meaning-change guidance in reference files read only when needed |
Setup and audit | in Claude Code, run in subagents: only their short report enters the session (other agents run them inline) |
How it runs
Hook | Does |
SessionStart | one line; snapshots existing issues so only new ones block later |
PostToolUse: Read | the claims a file holds, once per file |
PostToolUse: Write / Edit | drift in claim wording, outputs now out of date, suggestions for new files |
PostToolUse: Bash, media MCP tools | reminders to record generated media; tracked files that disappeared |
Stop | whole-project check; new problems and meaning changes block once (strict) or are reported (relaxed) |
Hooks never change claims. They write only git-ignored local files: session state, the activity log and the status cache (plus a debug log, only if you set CLAIMKEEPER_DEBUG). Errors are swallowed; claim tracking never breaks a session.
Storage and git
.claims/
config.json # only settings that differ from defaults (committed)
claims/<id>.json # one file per claim: meaning, history, places (committed)
outputs.json # generated outputs in tracked files (committed)
local/ # git-ignored: places/outputs in ignored files (e.g. out/*.mp4),
# personal settings, session state, activity.log, state.jsonPlaces in files git ignores go to
local/automatically, since nobody else has those files./claimkeeper:init --privatekeeps the whole store out of git.Only short anchor wording is stored (≤ 500 chars per place): the store grows with the number of claims, not the codebase.
Secrets are refused: anything that looks like an API key, token, private key or credential assignment is never stored;
.env*, key files andsecrets/folders are never tracked; generation metadata drops keys likeapi_key,token,authorization.
Security
A repo you clone or pull can ship its own .claims/ folder, so claimkeeper treats the store as untrusted input:
Paths are contained. Every path from the store or the agent must resolve inside the project: no absolute paths, no
.., and no symlinks or junctions that lead out of it. Anything else is reported asunsafe_pathand never read or written.The store and
.gitare never tracked, and the file walk never follows links. Excluded and protected paths are refused even when a store file names them.Store files are validated on load. A claim is always saved under its validated file name (never an id written inside the file), malformed entries are dropped, and exclude patterns are matched without backtracking.
Claim text can't impersonate claimkeeper. Descriptions, notes and wording are kept on one line, control and bidi characters are removed, and
[claimkeeper]inside stored text is neutralized.Secret protections can't be switched off. A project config can add exclusions, never remove the built-in ones (
.env*, key files,secrets/).The opt-in debug log redacts file contents, edits, commands and command output.
No runtime dependencies;
npm auditis clean. Regression tests for each of these live intest/security.test.ts.
Privacy
claimkeeper makes no network requests and sends nothing anywhere: no telemetry, no external services. Everything it stores stays in your project: the claims in .claims/ (committed with your repo unless you use --private) and your personal settings, session state and activity log in the git-ignored .claims/local/. It never stores secrets (see Security).
Tools (MCP)
Tool | Purpose |
| init, status, on/off, stop mode |
| create a claim and/or register its places (verified to exist; |
| list claims, or fetch claims with every place resolved live |
| only what needs attention, plus pending acknowledged items |
| claims in a file and outputs built from it; |
| new meaning ( |
|
|
CLI
Not on npm yet: run it from a checkout with node <checkout>/plugin/dist/cli.js.
node plugin/dist/cli.js init [--private]
node plugin/dist/cli.js on|off [--me]
node plugin/dist/cli.js check --ci # exit 1 if anything needs attention
node plugin/dist/cli.js get on-device
node plugin/dist/cli.js in README.md
node plugin/dist/cli.js log -n 50
node plugin/dist/cli.js install-git-hook # pre-commit check (uninstall-git-hook removes it)
node plugin/dist/cli.js status --root ../my-app # any command: --root or CLAIMKEEPER_ROOT picks the projectRules
Opt-in per project; off per user or per project.
A claim's one-sentence meaning is the source of truth; places store exact wording, verified on registration.
Rewording keeps the revision; a meaning change bumps it, marks unconfirmed places and outputs stale, and the agent tells you.
Only you decide what the product promises. A meaning change must say who decided (
user, oruser_requestwhen it follows from what you asked for) and why; the agent's own idea is refused until you agree. When places disagree and nothing you said settles it, the agent marks a conflict and asks instead of picking.git diff .claims/andclaimkeeper logshow who decided each change.Numbers are tied to code. A claim that states a number gets the constant that sets it as a code place, so changing
MAX_REPEATSflags the docs. Claims with numbers but no code place are listed byclaims_check.Web pages (a DEV post, a store listing) can't be checked by the tool: they're shown as "check by hand" with the wording to look for.
Media is never regenerated without your yes; "not now" is acknowledged and stays listed as pending. A conflict registered on purpose waits for your decision the same way, without blocking.
Missing files are reported, never fatal. Secrets are never stored.
Limits
It only knows registered claims and places. Keyword nudges,
claims_in scanand/claimkeeper:auditreduce the gap; they don't close it."Tell the user" and "never regenerate" are instructions to the agent, not locks.
Meaning vs rewording is the agent's judgment;
decided_bymakes the decision visible, but the agent still has to classify the change honestly.Audio and video are judged through what they were made from (scripts, captions, compositions). Claude can look at images and SVGs, not watch video.
Text built at runtime (
t('hero.title')) must be anchored where the literal string lives.Two sessions changing the same claim at the same moment: last write wins.
Development
npm install
npm test # builds plugin/dist, runs unit, hook, CLI and MCP tests
npm run validate # claude plugin validate (marketplace + plugin)src/core is the deterministic engine, src/mcp the stdio MCP server (no SDK), src/hooks the Claude Code hooks, plugin/ what ships (manifest, skills, agents, hooks.json, built dist/). Dependencies are pinned to exact versions.
License
MIT
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Fact-checks generated content against your sources of truth showing what to trust, change, & verify.
Source-backed prompting and coding-agent guidance with provenance and freshness checks.
Turn PRDs and product ideas into structured specs so coding agents build your intent, not theirs.
Runtime gate for AI agent claims. Catches sycophantic drift, circular self-validation, causal errors
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceTurn any link — video, PDF, screenshot, or article — into cached, timestamp-anchored understanding agents can query through lenses (explainer, build spec, teardown, design tokens, production blueprint). Every claim carries the exact second or page it came from.MIT
- AlicenseAqualityBmaintenanceAutonomous cross-source inconsistency and contradiction intelligence engine for AI agents. Detects specification drift, SemVer conflicts, and infrastructure mismatches across code, docs, and deployments.1210MIT
- AlicenseNot gradedqualityBmaintenanceEnables humans and AI coding agents to collaboratively write to a shared knowledge base with conflict-safe claims, code-anchored staleness detection, and synchronized human/agent documentation via MCP and REST.1AGPL 3.0
- AlicenseNot gradedqualityBmaintenanceEnables agents to verify README claims and apply structured edits that automatically revert if they fail the repository's gates.1Apache 2.0